The Role of Bile Salts in Diarrhoea Ofpatients with Ulcerative Colitis

Gut: first published as 10.1136/gut.12.8.632 on 1 August 1971. Downloaded from

Gut, 1971, 12, 632-635

The role of bile salts in diarrhoea of patients with ulcerative colitis

T. A. MIETTINEN From the Third Department of Medicine, University of Helsinki, Helsinki 29, Finland

SUMMARY Faecal bile salt elimination, which was determined in patients with ulcerative colitis, was mostly within normal limits, suggesting that in this disease water and electrolyte diarrhoea was hardly contributed to by excessive amounts of bile salts in the large bowel. In a therapeutic trial cholestyramine had no beneficial effect on the general condition, diarrhoea, or faecal composition, findings which further strengthen the view that bile salts play no role in the diarrhoea of ulcerative colitis, provided that the disease is limited to the large bowel. Faecal elimination of cholesterol as bile salts and neutral steroids tended initially to be higher than normal, but increased less than in controls during treatment with cholestyramine, particularly in the heavier patients. Serum cholesterol was low in the patients with ulcerative colitis and decreased by cholestyramine less (27 %) than in controls (39 %).

Recent studies have indicated that in patients with Material and Methods ileal dysfunction bile acids entering the colon in

augmented amounts can aggravate the diarrhoea The subjects studied were six controls (cases 1-6) http://gut.bmj.com/ (Hofmann and Poley, 1969) by enhancing colonic from among the laboratory personnel and five motility (Meyer and McEwan, 1948; Bergmann, patients (cases 7-11) with ulcerative colitis. Clinical 1952) and by stimulating electrolyte and water and laboratory data are presented in Table I. secretion into the colon and inhibiting their absorp- X-ray and sigmoidoscopic examinations indicated tion from the large bowel (Forth, Rummel, and that in two of the patients (cases 7 and 8) severe Glasner, 1966; Mekhjian, Phillips, and Hofmann, changes were present from the rectum up to the 1968; Phillips, Mekhjian, and Hofmann, 1970). transverse colon, while the caecum and ascending Binding of bile acids with cholestyramine (Rowe, colon appeared normal. In case 7, with the most on September 29, 2021 by guest. Protected copyright. 1967; Hofmann and Poley, 1969; Sickinger, 1969; severe x-ray findings, the patient later underwent Bahls, 1970; Miettinen and Lempinen, 1970) or proctocolectomy. Operation showed that the ascend- lignin (Eastwood and Girdwood, 1968) has been ing colon was affected but the terminal ileum was reported to alleviate diarrhoea in subjects with ileal intact. In cases 9 and 10 slight x-ray and sigmoid- resection and ileal bypass. Since bile salts are oscopic changes were observed, while in case 11 normally absorbed to a certain extent from the there were only sigmoidoscopic alterations, the colon in man (Mekhjian et al, 1968; Samuel, Saypol, radiograph being normal. Meilman, Mosbach, and Chafizadeh, 1968), All the patients and controls were placed on a damage to the colonic mucosa in ulcerative colitis low-cholesterol solid food diet (70-125 mg/day), the may inhibit this reabsorption. Bile acids may thus fat content being 80 g/2,400 kcal. Unabsorbable contribute to the diarrhoea and, by their possible Cr203 and sitosterol, 600 mg each divided into three toxic effect on intestinal mucosa (Shiner, 1969), daily doses, were given so that the respective worsen damage to the mucosa of the colon. There- corrections for faecal flow and degradation of fore, the role of bile acids in diarrhoea was evaluated cholesterol during intestinal transit (Grundy, Ahrens, in the present study by determining the faecal and Salen, 1968) could be made. Gas-liquid chrom- elimination of bile salts in patients with colonic atographic (GLC) analysis of the sitosterol showed lesions of varying severity and exploring the effect that it contained 85% of P-sitosterol and 15% of of cholestyramine treatment on clinical condition, other plant sterols. Measurement of dietary sterols faecal frequency, and faecal composition. indicated that the average total intake of ,- Received for publication 25 May 1971. sitosterol was 624 mg/day. Two stool collections, 632 Gut: first published as 10.1136/gut.12.8.632 on 1 August 1971. Downloaded from

The role ofbile salts in diarrhoea ofpatients with ulcerative colitis 633 each over two days, were made after five days on the was not attributable to excessive amounts of bile diet, Cr203, and fl-sitosterol. Thereafter chol- salts. In patient 7, who later underwent colectomy, estyramine (Cuemid), 32 g a day, was started for the values were slightly elevated, particularly when eight to 10 days and stool collections were con- expressed per kilogram of body weight. The sterol tinued. Serum cholesterol was determined (Pearson, balance, which showed a positive correlation with Stem, and McGawack, 1953) before and at the end body weight (r = 0-89 for patients; 0.66 for controls), of the cholestyramine period. tended to be higher in the patients than in the Faecal bile acids and neutral steroids and dietary controls, especially when expressed per kilogram of sterols were analysed by the gas liquid chromatog- body weight. Serum cholesterol was low, suggesting raphy methods (Grundy, Ahrens, and Miettinen, that the slightly increased elimination of cholesterol 1965; Miettinen, Ahrens, and Grundy, 1965), into the faeces was not sufficiently balanced by faecal fat according to van de Kamer, ten Bokkel enhanced cholesterol synthesis (Table II). Huinink, and Weyers (1949), and chromium as Cholestyramine clearly impaired the patient's presented by the methods of Bolin, King, and condition in case 8 both clinically and according to Klosterman (1952). Since no loss of f-sitosterol was faecal water and fat, while in the four remaining found during intestinal transit (recovery of dietary cases the status remained essentially unchanged. In plus administered ,-sitosterol in the faeces was cases 8-11 the cholestyramine treatment was followed complete in every case), the values are expressed in by administration of Salazopyrin and rectal pred- terms of Cr203. For this purpose the amount of nisone enemas, and immediate improvement was steroid or fat per mg of Cr203 in the faeces was recorded, all the symptoms disappearing within 10 multiplied by 600. Sterol balance, which in the days. Thus, cholestyramine has no favourable effect steady state is equal to cholesterol synthesis, was on ulcerative colitis limited to the colon and not given by the difference between dietary cholesterol extended to the terminal ileum. and faecal steroids (sum of bile acids and neutral Neither the average faecal mass, water, dry weight, steroids of cholesterol origin). nor fat were affected by cholestyramine in the five patients, while in the controls faecal mass was Results augmented, due to excessive water output. Bile acid

excretion and sterol balance values were markedly http://gut.bmj.com/ Data of Table I show that none of the patients had increased by cholestyramine in every patient but, in any gross steatorrhoea, though frequency of bowel contrast to controls, showed an inverse relation to movements and faecal mass (primarily due to excess- the body weight (Fig. 1). Thus, the average in- ive water) was markedly augmented. Faecal bile crement of faecal bile acid putput was significantly acid elimination was mostly normal (Table II), lower in the patients than in the controls. The fall however, and bore no significant relationship to the in serum cholesterol was correspondingly higher in faecal mass or water, so that the watery diarrhoea the latter (39 %) than in the former (27 %). on September 29, 2021 by guest. Protected copyright.

Case Age Sex Weight Relative No. of Faecal Components (g/day) No. (kg) Weight Daily Stools Total Weight Water Dry Matter Fat B D B D B D B D B D Control subjects 1-6 29 F & M 63 097 - - 120 228 92 155 29 73 1.9 2-4 Mean ± SE +2 ±5 ±0-02 ±24 ±16 ±22 ±14 ±3 ±4 ±03 ±0-6 Change' - - -- +107 ± 19' +64 ±17' +43 ± 2' +1-7 ± 09 Patients with ulcerative colitis 7 17 M 48 0-72 4-6 2-4 420 642 364 558 56 84 3-2 3-7 8 26 F 54 0.93 2-6 4-11 225 549 182 456 43 93 4-7 13-6 9 22 F 62 0.94 3-5 3-4 397 485 346 412 51 73 4.9 7-4 10 52 M 72 1-12 3-4 2-4 251 216 208 160 43 56 5.0 6-1 11 25 M 86 1.19 3-6 4-5 778 663 703 561 75. 72 8-0 7-8 7-11 28 64 0-98 -- 414 511 361 429 54 76 5.2 7-7 Mean ± SE +6 ±7 ±0-08 ±99' ±80' ±93' ±73' ±6' ±6 ±0-8 ±1.6' Change' - - - - +97 ±80 +68 ±76 +22 ±9 +2.5 ±116 Table I Clinical and laboratory data ofcontrol subjects and patienrts with ulcerative colitis treated with cholestyraminel The amount of unabsorbable resin (32 g/day) has not been subtracted from faecal weight or dry matter. B before and D = during cholestyramine treatment.'Mean difference (± SE) from pretreatment values. Statistically significant (p < 0O05) changes are indicated by ' and differences from the controls by '. Gut: first published as 10.1136/gut.12.8.632 on 1 August 1971. Downloaded from

634 T. A. Miettinen

Case Faecal Steroids (mg/day) Sterol Balance Serum No. Cholesterol Bile Neutral Total (mg/l00 mi) Acids Steroids Steroids mg/day mg/kg/day Control subjects i-6 232 631 863 -770 -12.3 209 Mean + SE ±45 ±108 ±117 ±110 ±14 ±11 Changel +1,776 +42 +1,818 -1,818 -28.9 -81 ±208' ±57 ±255' t±255' i±2.4' ±10' Patients with ulcerative colitis 7 463 427 890 -796 -166 115 8 159 936 1,095 -1,012 -18-7 184 9 288 830 1,118 -1,028 -16.8 153 10 238 927 1,165 -1,071 -14.9 150 11 353 950 1,303 -1,178 -13.7 180 7-11 300 814 1,114 -1,107 -16-1 157 Mean ± SE ±52 ±99 ±67 ±62 ±0 8' ±15' Change' +1,055 +95 +1,150 -1,150 -19.9 -42 ± 160"' ±59 ±209':±209' ±5A2' ±82"' Table II Faecal steroids, sterol balance, and serum cholesterol in control subjects and patients with ulcerative colitis treated with cholestyramine 'Mean change (± SE) caused by cholestyramine. Statisticaily significant (p < 005) changes indicated by ' and differences from the controls by'

80 concentrations not only inhibit electrolyte and water absorption from the large bowel but also stimulate 0 the secretion of sodium and water into the intestine nfl (Forth et al, 1966; Mekhjian et al, 1968; Phillips 75 - 0 et al, 1970). These mechanisms appear to contribute to the diarrhoea in ileal dysfunction so that binding of bile acids with cholestyramine decreases faecal http://gut.bmj.com/ 00 weight, water content, and frequency (Rowe, 1967; 60- Hofman and Poley, 1969; Sickinger, 1969; Bahls, 10 a 1970; Miettinen and Lempinen, 1970). The results 0 of the present study showed that in patients with 0 ulcerative colitis, limited to the large bowel, there * is no excessive faecal loss of bile salts, so that +0.5 +1.3 +2.9 cholerrhoeic diarrhoea does not exist. The absence of large amounts of bile salts in the on September 29, 2021 by guest. Protected copyright. A Faecal steroids (g/day) colon obviously explains the failure of choles- Fig. 1 Correlation ofbody weight with faecal steroid tyramine to alleviate the diarrhoea by binding bile increment in control subjects (0) and in patients with acids. This also indicates that bile salts, known to ulcerative colitis (0). injure intestinal cells in vivo at high concentrations (Shiner, 1969), have hardly any additional damaging effect on the diseased colonic mucosa in ulcerative Discussion colitis. But in the cases in which the inflammatory and ulcerative processes extend to the terminal Diarrhoea of ulcerative colitis depends on enhanced ileum ('back-wash' ileitis) ileal dysfunction may exudation of electrolytes and water through the develop. Under these conditions large amounts of damaged mucosa and impaired absorption ofsodium bile salts may escape into the colon, contributing to and water from the ileal contents of the large bowel the diarrhoea and perhaps worsening the mucosal (Duthie, Watts, deDombal, and Goligher, 1964; ulceration through their cytotoxicity. In regional Harris and Shields, 1970). If bile salt absorption, enteritis enhanced faecal bile salt excretion has known to occur in the colon to some extent actually been found in many but not all cases (Mekhjian et al, 1968; Samuel et al, 1968), were (Stanley and Nemchausky, 1967; Meihoff and Kern, similarly diminished, then cholerrhoeic diarrhoea 1968; Fiasse, Eyssen, Dive, Harvengt, Kestens, might contribute to the excessive water and and Nagant de Deuxchaisnes, 1970). electrolyte loss found in this disease. Bile salts, The reason for the small increase in bile salt particularly free dihydroxy bile acids, at high enough output during cholestyramine treatment in the Gut: first published as 10.1136/gut.12.8.632 on 1 August 1971. Downloaded from

The role ofbile salts in diarrhoea ofpatients with ulcerative colitis 635 of bile acids, lipids and calcium in Crohn's disease with and colitis patients may be due to effective reabsorption without resection. In 4th World Congress of Gastroenterology, of bile salts or limited hepatic capacity to enhance Copenhagen, 1970, (Abstracts), p. 354. bile salt synthesis, particularly in the two slightly Forth, W., Rummel, W., and Glasner, H. (1966). Zur resorptions- hemmenden Wirkung von Gallensauren. Naunyn-Schmiede- obese subjects. The limited synthesis may have been berg's Arch. exp. Path. Pharmak., 254, 364-380. caused by anorexia and weight loss, which are almost Grundy, S. M., Ahrens, E. H. Jr., and Miettinen, T. A. (1965). Quantitative isolation and gas-liquid chromatographic analysis invariably part of the clinical picture of active of total fecal bile acids. J. Lipid Res., 6, 397-410. ulcerative colitis, or by associated liver disorder, Grundy, S. M., Ahrens, E. H. Jr., and Salen, G. (1968). Dietary 8- which is frequently found in this disease (cf. Wright, sitosterol as an internal standard to correct for cholesterol losses in sterol balance studies. J. Lipid Res., 9, 374-387. 1970). No laboratory signs of liver damage were Harris, J., and Shields, R. (1970). Absorption and secretion of water seen; liver biopsies were not performed. Peri- and electrolytes by the intact human colon in diffuse un- treated proctocolitis. Gut, 11, 27-33. cholangitis and fatty change are often found in Hofmann, A. F., and Poley, J. R. (1969). Cholestyramine treatment liver biopsies in ulcerative colitis, even when the of diarrhea associated with ileal resection. New Engi. J. Med., 281, 397-402. liver function tests are normal (cf. Wright, 1970). van de Kamer, J. H., ten Bokkel Huinink, H., and Weyers, H. A. All the patients had lost weight before admission to (1949). Rapid method for the determination of fat in feces. hospital and tended to do so even during their stay J. biol. Chem., 177, 347-355. Meihoff, W. E., and Kern, F. Jr. (1968). Bile salt malabsorption in in hospital, despite an apparently sufficient caloric regional ileitis, ileal resection, and mannitol-induced diarrhea. intake. Faecal bile acid excretion and bile salt J. clin. Invest., 47, 261-267. Mekhjian, H. S., Phillips, S. F., and Hofmann, A. F. (1968). Con- synthesis are actually reduced in man during caloric jugated bile salts block water and electrolyte transport by the restriction not only under basal conditions but human colon. Gastroenterology, 54, 1256. Meyer, A. E., and McEwen, J. P. (1948). Bile acids and their choline apparently also during cholestyramine treatment salts applied to the inner surface of the isolated colon and (Miettinen, 1968). However, the fact that no marked ileum of the guinea pig. Amer. J. Phyiiol., 153, 386-392. steatorrhoea developed during cholestyramine treat- Miettinen, T. A. (1968). Fecal steroid excretion during weight reduc- tion in obese patients with hyperlipidemia. Clin. chim. Acta, ment in the patients with ulcerative colitis suggests 19, 341-344. that the intestinal bile salt concentration was Miettinen, T. A., and Lempinen, M. (1970). Ileal by-pass operation in familial hypercholesterolemia. (Abstr.) Scand. J. clin. Lab. maintained high enough for fat absorption by Invest., 25, Suppl. 113, 55. enhanced hepatic synthesis of bile salts. Miettinen, T. A., Ahrens, E. H. Jr., and Grundy, S. M. (1965). Quantitative isolation and gas-liquid chromatographic analysis of total dietary and fecal neutral steroids. J. Lipid Res., 6, 411- Skilful technical and secretarial assistance given by 424. http://gut.bmj.com/ Mrs E. Gustafsson, Mrs P. Hoffstrom, and Mrs U Pearson, S., Stern. S., and McGawack, T. H. (1953). A rapid, accurate method for the determination of total cholesterol in serum. Kaski is acknowledged. The study has been per- Analyt. Chem., 25, 813-814. formed under a contract with the Association of Phillips, S. F., Mekhjian, H. S., and Hofmann, A. F. (Copenhagen 1970). Effect of free and conjugated bile acids on electrolyte Finnish Life Assurance Companies. and water absorption by the human colon. In 4th World Congress of Gastroenterology Copenhagen, 1970, (Abstracts), References pp. 563-564. Rowe, G. G. (1967). Control of tenesmus and diarrhea by chol-

Bahls, G. (1970). Chologene Diarrhoen nach Ileumresektion und estyramine administration. Gastroenterology, 53, 1006. on September 29, 2021 by guest. Protected copyright. ihre Behandlung mit Cholestyramin. Z. Gastroent., 8, 41-43. Samuel, P., Saypol, G. M., Meilman, E., Mosbach, E. H., and Bergmann, M. (1952). Ueber die peristaltische Wirksamkeit der Chafizadeh, M. (1968). Absorption of bile acids from the large Gallensauren und ihre klinische Anwendung. Wien klin. bowel in man. J. clin. Invest., 47, 2070-2078. Wschr., 64, 704.707. Shiner, M. (1969). Effect of bile acids on the small intestinal mucosa Bolin, D. W., King, R. P., and Klosterman, E. W. (1952). A simplified in man and rats: a light and electron microscope study. In method for the determination of chromic oxide (Cr0,O) when Bile Salt Metabolism, edited by L. Schiff, J. B. Carey, Jr. used as an index substance. Science, 116, 634.635. and J. M. Dietschy, pp. 41-55. Thomas, Springfield, Illiois. Duthie, H. L., Watts, J. M., deDombal, F. T., and Goligher, J. C. Sickinger, K. (1969) Die Behandlung der chologenen Diarrhoe und (1964). Serum electrolytes and colonic transfer of water and Steatorrhoe des enteralen Gallensaureverlustsyndroms mit electrolytes in chronic ulcerative colitis. Gastroenterology, 47, Cholestyramin und mittelkettigen Triglyceriden. Dtsch. med. 525-530. Wschr., 94, 1151-1157. Eastwood, M. A., and Girdwood, R. H. (1968). Lignin: a bile-salt Stanley, M. M., and Nemchausky, B. (1967). Fecal C"4-bile acid sequestrating agent. Lancet, 2, 1170.1172. excretion in normal subjects and patients with steroid-wasting Fiasse, R., Eyssen, R., Dive, C., Harvengt, C., Kestens, P. J., and syndromes secondary to ileal dysfunction. J. Lab. clin. Med., Nagant de Deuxchaisnes, C. (Copenhagen 1970). Metabolism 70, 627-639. Wright, R. (1970). Ulcerative colitis. Gastroenterology, 58, 875-897.